EP3678255B1 - Method for calibrating a family of lithium-ion battery elements, charging method, associated computer program product and charging device - Google Patents

Description

TECHNICAL AREA

The present invention relates to a method for calibrating a family of Li-ion battery cells. Also provided are a charging method, a computer program product and a charging device.

The invention applies to the field of batteries, more particularly to the field of lithium-ion batteries.

STATE OF PRIOR ART

Lithium-ion batteries, also known as Li-ion, generally have a higher specific energy than other types of batteries. As a result, Li-ion batteries constitute an advantageous alternative for the storage of electrical energy.

Today, Li-ion batteries comprise at least one battery element, each battery element comprising a so-called “positive” electrode (or cathode) and a so-called “negative” electrode (or anode), separated by an electrolyte. The cathode is generally metallic, for example made of a compound of the transition metal oxide type or of the lithium phosphate type comprising lithium. In addition, the anode contains graphite particles.

Li-ion battery cells which include an anode of a similar nature (that is to say an anode with in particular the same morphology of the graphite particles, the same composition, the same binder and the same porosity), a similar cathode (that is to say a cathode having in particular the same composition: the same binder, the same conductive additive, the same porosity) and a similar electrolyte (that is to say an electrolyte with, in particular, the same ionic conductivity) define a “family”, within the meaning of the present invention.

In operation, when charging a Li-ion battery, Li ⁺ lithium ions migrate from the cathode of each battery cell to insert themselves into the graphite of the corresponding anode. Reverse migration occurs when the battery discharges.

Although having a high energy mass density, such batteries generally see their performance degrade over time.

Such degradation is due to the formation of a metallic lithium deposit on the surface of the graphite particles of the anode. Such a phenomenon generally occurs when charging the battery, and is commonly referred to by the English expression “ lithium plating”.

When such a deposit comes into contact with the electrolyte, the lithium generally oxidizes to form compounds such as lithium carbonates (ROCO ₂ Li) or inorganic compounds (Li ₂ CO ₃ , LiF). The lithium then becomes inactive from an electrochemical point of view, and is no longer able to participate in the redox reactions taking place at the electrodes.

In addition, such a deposit is also likely to become detached from the graphite particles of the anode. In this case, the lithium in metallic form, although active from an electrochemical point of view, is no longer electronically connected to the electrode and is therefore no longer likely to be exchanged between the electrodes to participate in transport. of electric current in the battery cells. This loss of exchangeable lithium leads to a progressive and irreversible drop in battery capacity.

The phenomenon of lithium deposition contributes to the loss of battery capacity, thereby reducing its lifespan.

The document US 2005/0233220 A1 describes a Li-ion battery manufactured so as to be less subject to metallic lithium deposition phenomena than usual Li-ion batteries.

However, such a battery does not give complete satisfaction.

Indeed, such a battery also sees its performance degrade over time due to the formation of metallic lithium deposits at the anode.

He noted that the documents FROM 10 2013 007011 A1 , US 2018/123354 A1 And FROM 10 2016 007479 A1 also provide teachings relevant to the present invention.

In fact, the document FROM 10 2013 007011 A1 described with reference to his figure 1 , a method of charging a battery comprising at least one Li-ion battery cell. The document US 2018/123354 A1 describes, with reference to its figure 9, a method for calibrating a family of Li-ion battery elements. Regarding the document FROM 10 2016 007479 A1 also describes a method for calibrating a family of Li-ion battery cells.

An aim of the invention is to propose a calibration method which makes it possible to extend the lifespan of Li-ion batteries.

STATEMENT OF THE INVENTION

• fourniture d'un élément de batterie de la famille, l'élément de batterie présentant une capacité surfacique initiale donnée correspondante ;
• au moins une évaluation d'un état de santé courant de l'élément de batterie fourni ;
• pour l'élément de batterie fourni, mise en oeuvre d'une pluralité de cycles successifs comprenant chacun une phase de charge et une phase de décharge, au moins deux phases de charge étant réalisées à des régimes de charge différents ;
• pour chaque phase de charge :
  • mesure d'un potentiel électrique d'une anode de l'élément de batterie fourni ;
  • calcul de la valeur d'une grandeur caractéristique prédéterminée de l'élément de batterie ;
  • détermination d'une valeur limite de la grandeur caractéristique prédéterminée pour laquelle le potentiel de l'anode devient inférieur ou égal à un seuil prédéterminé ;
  • enregistrement, dans une mémoire, de la valeur limite pour ladite famille ladite capacité surfacique initiale, le régime de charge de ladite phase de charge et l'état de santé évalué.

To this end, the subject of the invention is a process of the aforementioned type, comprising the steps of:

• providing a battery cell of the family, the battery cell having a corresponding given initial surface capacity;
• at least one assessment of a current health state of the supplied battery cell;
• for the supplied battery cell, implementation of a plurality of successive cycles each comprising a charging phase and a discharging phase, at least two charging phases being carried out at different charging regimes;
• for each charging phase:
  • measuring an electrical potential of an anode of the supplied battery cell;
  • calculation of the value of a predetermined characteristic quantity of the battery cell;
  • determination of a limit value of the predetermined characteristic quantity for which the potential of the anode becomes less than or equal to a predetermined threshold;
  • recording, in a memory, the limit value for said family, said initial surface capacity, the charging regime of said charging phase and the assessed state of health.

Indeed, the phenomenon of metallic lithium deposition mainly occurs when the anode potential becomes negative.

Such a method makes it possible to determine the limit value of the predetermined characteristic quantity, for which the anode potential is still positive or zero. The use of such a limit value gives the possibility, when charging the battery cell, to interrupt the charging of the battery cell when the current value of the predetermined characteristic quantity reaches the limit value, or to reduce the charging regime in favor of a lower charging regime, for which the limit value of the predetermined characteristic quantity is higher than the limit value associated with the current charging regime. In addition, the evaluation of the state of health of the battery cell allows the degradation of the performance of the battery cell to be taken into account over time.

• chaque évaluation de l'état de santé courant de l'élément de batterie fourni est mise en oeuvre au cours d'un cycle correspondant ;
• le procédé d'étalonnage comporte le calcul, par régression, d'un modèle mathématique liant la valeur limite de la grandeur caractéristique prédéterminée au régime de charge ;
• le modèle mathématique lie la valeur limite de la grandeur caractéristique prédéterminée au régime de charge, à la capacité surfacique initiale de l'élément de batterie fourni et à l'état de santé de l'élément de batterie fourni ;
• le modèle mathématique est un modèle quadratique de la forme : $${\mathrm{VAL}}_{\lim }={\mathrm{a}}_0+{\mathrm{a}}_1{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0+{\mathrm{a}}_2\log \left(\mathrm{C}\right)+{\mathrm{a}}_3\mathrm{H}+{\mathrm{a}}_{12}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\log \left(\mathrm{C}\right)+{\mathrm{a}}_{13}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\mathrm{H}+{\mathrm{a}}_{23}\log \left(\mathrm{C}\right).\mathrm{H}+{\mathrm{a}}_{11}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0^2+{\mathrm{a}}_{22}\mathrm{<figure-callout\; id="2"\; label="log\; C"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">log</figure-callout>}{\left(\mathrm{C}\right)}^{}{\left(\mathrm{}\right)}^2+{\mathrm{a}}_{33}{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">H</figure-callout>}}^2$$
  
  • où VAL_lim est la valeur limite de la grandeur caractéristique prédéterminée ;
  • Q₀ est la capacité surfacique initiale de l'élément de batterie ;
  • H est l'état de santé de l'élément de batterie ;
  • C est le régime de charge ;
  • log est l'opérateur logarithme décimal ; et
  • a₀, a₁, a₂, a₃, a₁₃, a₂₃, a₁₁, a₂₂ et a₃₃ des coefficients réels ;
• pour chaque cycle, l'élément de batterie est maintenu à température constante, au moins deux phases de charge étant réalisées à des températures différentes, la valeur limite enregistrée pour chaque phase de charge étant associée à la température correspondante, le modèle mathématique calculé liant la valeur limite de la grandeur caractéristique prédéterminée au régime de charge, à la capacité surfacique initiale de l'élément de batterie fourni, à l'état de santé de l'élément de batterie fourni ainsi qu'à la température de l'élément de batterie fourni ;
• le potentiel de l'anode est calculé à partir de la différence de potentiel entre l'anode et une électrode de référence ;
• l'électrode de référence est une électrode Li⁺/Li, la valeur limite de la grandeur caractéristique prédéterminée étant atteinte lorsque la différence de potentiel entre l'anode et l'électrode de référence est nulle ;
• la grandeur caractéristique prédéterminée est un état de charge de l'élément de batterie, ou une tension entre une cathode et l'anode de l'élément de batterie.

According to other advantageous aspects of the invention, the process comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations:

• each evaluation of the current state of health of the supplied battery cell is carried out during a corresponding cycle;
• the calibration method comprises the calculation, by regression, of a mathematical model linking the limit value of the predetermined characteristic quantity to the load regime;
• the mathematical model links the limit value of the predetermined characteristic quantity to the charging regime, to the initial surface capacity of the supplied battery cell and to the state of health of the supplied battery cell;
• the mathematical model is a quadratic model of the form: $${\mathrm{VAL}}_{\mathrm{lime}}={\mathrm{has}}_0+{\mathrm{has}}_1{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0+{\mathrm{has}}_2\log \left(\mathrm{VS}\right)+{\mathrm{has}}_3\mathrm{H}+{\mathrm{has}}_{12}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\log \left(\mathrm{VS}\right)+{\mathrm{has}}_{13}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\mathrm{H}+{\mathrm{has}}_{23}\log \left(\mathrm{VS}\right).\mathrm{H}+{\mathrm{has}}_{11}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0^2+{\mathrm{has}}_{22}\log {\left(\mathrm{VS}\right)}^2+{\mathrm{has}}_{33}{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">H</figure-callout>}}^2$$
  
  • where VAL _lim is the limit value of the predetermined characteristic quantity;
  • Q ₀ is the initial surface capacity of the battery cell;
  • H is the health status of the battery cell;
  • This is the charging regime;
  • log is the decimal logarithm operator; And
  • a ₀ , a ₁ , a ₂ , a ₃ , a ₁₃ , a ₂₃ , a ₁₁ , a ₂₂ and a ₃₃ real coefficients;
• for each cycle, the battery cell is maintained at constant temperature, at least two charging phases being carried out at different temperatures, the limit value recorded for each charging phase being associated with the corresponding temperature, the calculated mathematical model linking the limit value of the predetermined characteristic quantity at the charging rate, the initial surface capacity of the supplied battery cell, the state of health of the supplied battery cell as well as the temperature of the battery cell provided ;
• the anode potential is calculated from the potential difference between the anode and a reference electrode;
• the reference electrode is a Li ⁺ /Li electrode, the limit value of the predetermined characteristic quantity being reached when the potential difference between the anode and the reference electrode is zero;
• the predetermined characteristic quantity is a state of charge of the battery cell, or a voltage between a cathode and the anode of the battery cell.

• fourniture d'une batterie comprenant au moins un élément de batterie Li-ion appartenant à une famille d'éléments de batterie Li-ion et présentant une capacité surfacique initiale donnée ;
• évaluation d'un état de santé de chaque élément de batterie de la batterie ;
• alimentation de la batterie fournie en énergie électrique, à un régime de charge donné ;
• pour chaque élément de batterie, calcul d'une valeur limite d'une grandeur caractéristique prédéterminée correspondante, pour ladite famille, ladite capacité surfacique initiale, ledit état de santé et ledit régime de charge ;
• pour chaque élément de batterie, calcul de la valeur de la grandeur caractéristique prédéterminée ; et
• détection d'un état limite lorsque, pour au moins un élément de batterie, la valeur de la grandeur caractéristique prédéterminée atteint une fraction prédéterminée de la valeur limite correspondante.

Furthermore, the subject of the invention is a charging method comprising the steps of:

• provision of a battery comprising at least one Li-ion battery element belonging to a family of Li-ion battery elements and having a given initial surface capacity;
• evaluating a health status of each battery cell of the battery;
• supply of the battery supplied with electrical energy, at a given charging regime;
• for each battery element, calculation of a limit value of a corresponding predetermined characteristic quantity, for said family, said initial surface capacity, said state of health and said charging regime;
• for each battery element, calculation of the value of the predetermined characteristic quantity; And
• detection of a limit state when, for at least one battery element, the value of the predetermined characteristic quantity reaches a predetermined fraction of the corresponding limit value.

The limit value is obtained by applying the calibration process as defined above.

• la méthode de charge comporte, en outre, la réduction du régime de charge en cas de détection de l'état limite.

According to another advantageous aspect of the invention, the charging method comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations:

• the charging method also includes reducing the charging regime in the event of detection of the limit state.

Furthermore, the invention relates to a computer program product comprising program code instructions which, when executed by a computer, implement the charging method as defined above.

• un circuit d'alimentation configuré pour transmettre de l'énergie électrique à une batterie Li-ion, à un régime de charge donné, depuis une source d'énergie électrique ;
• un calculateur configuré pour :
  • évaluer un état de santé de chaque élément de batterie de la batterie ;
  • calculer la valeur d'une grandeur caractéristique prédéterminée relative à au moins un élément de batterie de la batterie au cours du temps ;
  • détecter un état limite lorsque la valeur de la grandeur caractéristique prédéterminée relative à au moins un élément de batterie atteint une fraction prédéterminée d'une valeur limite, la valeur limite étant fonction d'une famille et d'une capacité surfacique initiale de l'élément de batterie, de l'état de santé évalué et du régime de charge.

Furthermore, the invention relates to a charging device comprising:

• a power circuit configured to transmit electrical energy to a Li-ion battery, at a given charging rate, from an electrical energy source;
• a calculator configured to:
  • evaluate a health status of each battery cell of the battery;
  • calculate the value of a predetermined characteristic quantity relating to at least one battery cell of the battery over time;
  • detect a limit state when the value of the predetermined characteristic quantity relating to at least one battery element reaches a predetermined fraction of a limit value, the limit value being a function of a family and an initial surface capacity of the element battery, assessed state of health and charging regime.

According to another advantageous aspect of the invention, the computer is, in addition, configured to control the reduction of the charging speed in the event of detection of the limit state.

BRIEF DESCRIPTION OF THE DRAWINGS

• la figure 1 est une représentation schématique d'un élément de batterie lithium-ion ;
• la figure 2 est un organigramme illustrant un procédé d'étalonnage selon l'invention ;
• la figure 3 est un organigramme illustrant une étape cyclique du procédé d'étalonnage selon l'invention ;
• la figure 4 est un graphique représentant l'évolution d'un potentiel d'anode d'un élément de batterie Li-ion en fonction de l'état de charge, pour un régime de charge et une capacité surfacique initiale donnés de l'élément de batterie, et pour quatre valeurs distinctes de l'état de santé ;
• la figure 5 est une représentation schématique d'un dispositif de charge selon l'invention ; et
• la figure 6 est un organigramme illustrant le fonctionnement du dispositif de charge de la figure 5.

The invention will be better understood with the help of the description which follows, given solely by way of non-limiting example and made with reference to the appended drawings in which:

• there figure 1 is a schematic representation of a lithium-ion battery cell;
• there figure 2 is a flowchart illustrating a calibration method according to the invention;
• there Figure 3 is a flowchart illustrating a cyclical step of the calibration process according to the invention;
• there figure 4 is a graph representing the evolution of an anode potential of a Li-ion battery cell as a function of the state of charge, for a given charging regime and a given initial surface capacity of the battery cell, and for four distinct values of health status;
• there figure 5 is a schematic representation of a charging device according to the invention; And
• there Figure 6 is a flowchart illustrating the operation of the charging device of the figure 5 .

DETAILED DESCRIPTION OF PARTICULAR EMBODIMENTS

A battery element 2 is schematically represented on the figure 1 . The battery element 2 is such that a battery is obtained by the series and/or parallel assembly of a plurality of battery elements 2. More precisely, the battery element 2 is an element of a Li-ion battery.

Conventionally, the battery element 2 comprises a so-called positive electrode, also called "cathode" 4, and a so-called negative electrode, also called "anode" 6. The cathode 4 and the anode 6 are arranged at a distance from each other. one from the other, the space between the electrodes 4, 6 being occupied by an electrolyte 8.

The battery cell 2 has a surface capacity Q, generally expressed in milliampere hours per square centimeter (mAh/cm ² ).

By “surface capacity” is meant the equivalent capacity of the active material per unit surface, that is to say an equivalent capacity obtained from the capacity representative of the performance of the electrochemically active material (in mAh/g ) within the electrodes, relative to the surface of the electrodes and multiplied by the quantity of active material actually present in the electrodes (in grams).

Such a surface capacity varies over time and, more specifically, decreases, from an initial maximum value, called initial surface capacity Q ₀ , as a state of health H of the battery cell increases. 2 is deteriorating.

Such a state of health H (in English, “ state of health ”) constitutes an indicator of the aging of the battery cell 2. Such aging comes, for example, from a deposit of metallic lithium at the anode. 6.

Means of evaluating the state of health H are, for example, described in the document t “Towards a smarter battery management system: A critical review on battery state of health monitoring methods”, Rui Xiong et al., Journal of Power Sources, Volume 405, Pages 18-29 (2018 ). We can cite, as an example, the coulometric counting method which consists, after a complete charge, of integrating the current throughout the discharge and thus obtaining the discharge capacity at cycle n corresponding. The state of health indicator can then be defined as being the ratio between this discharge capacity at cycle n and the initial capacity of the accumulator.

The battery cell 2 is such that, before the first charge of the battery cell 2, the cathode 4 contains lithium. For example, the cathode 4 is made of a compound comprising lithium, such as LiNi _1/3 Mn _1/3 Co _1/3 O ₂ .

Preferably, the anode 6 comprises graphite particles.

Preferably, the electrolyte 8 is a solution of a lithium salt dissolved in a mixture of alkylcarbonates. For example, the lithium salt is lithium hexafluorophosphate LiPF ₆ . For example, the alkyl carbonate mixture is a mixture of ethylene carbonate, ethyl methyl carbonate and dimethyl carbonate.

The calibration method of the family to which the battery element 2 belongs is described with reference to the figure 2 .

A plurality of battery cells 2 of the same family is provided. At least two battery elements 2 of the family have a different initial surface capacity Q ₀ and a different state of health H.

During an initialization step 10, a battery cell 2 is chosen from the plurality of battery cells 2. The state of health H of the chosen battery cell 2 is evaluated, i.e. -say calculated. Alternatively, the state of health H is evaluated during each cyclic step 12 described subsequently.

The chosen battery cell 2 is initially discharged.

The chosen battery cell 2 is then connected to a calibration device (not shown).

The step following the initialization step 10 is a cyclic step 12, as described later.

Advantageously, during each cyclic step 12, the calibration device keeps the temperature of the battery cell 2 constant.

Preferably, the calibration device is configured to implement cyclic step 12 following a predetermined number of iterations. In this case, for at least two distinct implementations of cyclic step 12, the corresponding temperatures are preferably different.

Furthermore, in this same case where the calibration device is configured to implement the cyclic step 12 following a predetermined number of iterations, the step following each cyclic step 12 is a test step 14.

During the test step 14, the calibration device determines whether the number of times during which the cyclic step 12 was implemented has reached the predetermined number of iterations.

If, during the test step 14, the calibration device determines that the cyclic step 12 has been implemented a number of times strictly less than the predetermined number of iterations, then the step following step test step 14 is a new cyclic step 12. If, during the test step 14, the calibration device determines that the cyclic step 12 has been implemented a number of times equal to the predetermined number of iterations , then the step following the test step 14 is a calculation step 16.

During any given cyclic step 12, the calibration device subjects the battery cell 2 to a predetermined number of cycles each comprising a charging phase 12.1 and a discharge phase 12.2 of the battery cell 2, as illustrated by the Figure 3 .

At the end of each cycle, the calibration device determines, during a control phase 12.3, whether the number of cycles carried out has reached the predetermined number of cycles. If the number of cycles carried out is strictly less than the predetermined number of cycles, then a new cycle comprising a charging phase 12.1 and a discharging phase 12.2 is implemented.

If the number of cycles performed reaches the predetermined number of cycles, then cyclic step 12 is followed by test step 14.

During each charging phase 12.1, the calibration device supplies the battery cell 2 with electrical energy at a predetermined constant charging rate.

By “charging regime” is meant, within the meaning of the present invention, an electric current calculated from the maximum constant current that a battery or that a battery cell 2 is capable of delivering in one hour. For example, for a battery cell capable of discharging 1000 mA (milliampere) in one hour, the charging regime 1C corresponds to a current of 1000 mA, the charging regime C/10 corresponds to an electric current of 100 mA, the 3C charging regime corresponds to an electric current of 3000 mA, etc.

Preferably, for each charging phase 12.1 of the same cyclic step 12, the calibration device charges the battery cell 2 at a charging rate higher than the charging rate associated with the previous charging phase 12.1. For example, the charging regime is successively C/10, then C/2, then 0.75C, then 1C, then 1.25C, then 1.5C, then 2C, then 3C, then 5C.

In addition, the calibration device measures, during each charging phase 12.1 of cyclic step 12, the voltage across the battery cell 2.

In addition, during each charging phase 12.1, the calibration device measures the potential of the anode 6 of the battery element 2. Such potential of the anode 6 generally decreases with the increase in a state of charge S of battery cell 2, as defined later.

Preferably, the calibration device calculates the potential of the anode 6 from a measurement of the potential difference between the anode 6 and a reference electrode presenting a fixed potential over time, for example an electrode at lithium Li ⁺ /Li classically known.

In addition, during each charging phase 12.1, the calibration device calculates the state of charge S of the battery cell 2.

• où S(t) est l'état de charge à un instant t donné ;
• q(t) est la capacité de l'élément de batterie 2 à l'instant t considéré ; et
• q_max est la capacité maximale de l'élément de batterie 2 à l'issue de la phase de charge associée au régime de charge le plus faible de l'étape cyclique 12.

The state of charge S at a given instant of cyclic step 12 is a percentage defined as the result of dividing the capacity of the battery element 2 at the instant considered, by the maximum capacity of the element of battery 2 at the end of the charging phase associated with the lowest charging regime of cyclic step 12, multiplied by 100, as shown by formula (1): $$\mathrm{S}\left(\mathrm{t}\right)=100\frac{\mathrm{q}\left(\mathrm{t}\right)}{{\mathrm{q}}_{\max }}$$

• where S(t) is the state of charge at a given time t;
• q(t) is the capacity of battery cell 2 at the time t considered; And
• q _max is the maximum capacity of battery cell 2 at the end of the charging phase associated with the lowest charging regime of cyclic step 12.

For example, in the case where the lowest charging regime during the cyclic step 12 is C/10, q _max is the maximum capacity of the battery cell 2 at the end of the charging phase associated with the regime charge C/10.

The capacity q(t) of the battery element 2 at time t is, for example, taken equal to the integral, between a predetermined initial time and the current time t, of the measured value of the current circulating at through battery cell 2. The value of the capacity q(t) is positive. The predetermined initial instant is, for example, the instant at which the charging phase 12.1 begins.

The state of charge S increases during charging phase 12.1.

The state of charge S constitutes a predetermined characteristic quantity of the battery element 2, the value of which is representative of the value of the electric potential of the anode 6.

During each charging phase 12.1, the calibration device also determines a limit value VAL _lim of the state of charge S, for which the potential difference between the anode 6 and the reference electrode becomes less than or equal at a predetermined threshold. Advantageously, in the case where the reference electrode is a lithium reference electrode, the limit value VAL _lim corresponds to a zero potential difference between the anode 6 and the reference electrode.

The calibration device records the limit value VAL _lim , the limit value VAL _lim being associated with the family of the chosen battery element 2, with the initial surface capacity Q ₀ of the chosen battery element 2, with the state of health H of the chosen battery element 2 and at the charging regime of the charging phase 12.1 considered. Preferably, if the cyclic steps 12 are implemented at different respective temperatures, the recorded limit value VAL _lim is also associated with the respective temperature at which the chosen battery element 2 is maintained during each cyclic step 12.

As an example, the graph of the figure 4 illustrates the evolution of the anode potential of a Li-ion battery cell 2 as a function of its state of charge S, for the same charging regime and the same initial surface capacity Q ₀ given for the battery cell 2 , for four distinct values of the state of health H, the temperature here being constant and identical for all the measurements.

Curve 41 (dotted lines) illustrates the evolution of the anode potential as a function of the state of charge S, when the battery element 2 has a state of health of 100%.

Curve 42 (thick solid line) illustrates the evolution of the anode potential as a function of the state of charge S, when the battery element 2 has a state of health of 90%.

Curve 43 (thin solid line) illustrates the evolution of the anode potential as a function of the state of charge S, when the battery element 2 has a state of health of 80%.

Curve 44 (dashed lines) illustrates the evolution of the anode potential as a function of the state of charge S, when the battery element 2 has a state of health of 60%.

It appears that, for a given temperature and initial surface capacity Q ₀ , the limit value VAL _lim of the state of charge S, that is to say the value of the state of charge S for which the potential d The anode cancels out, is all the weaker as the state of health H of the battery element 2 is low.

The calibration device ends each charging phase 12.1 when a predetermined condition is met. For example, the calibration device ends each charging phase 12.1 when the voltage across the battery cell 2 reaches a predetermined maximum voltage.

During each discharge phase 12.2, the calibration device takes electrical energy from the battery cell 2, at a predetermined charging rate.

For example, the calibration device takes energy from the battery cell 2 at an identical charging rate for all the discharge phases 12.2 of the cyclic step 12, for example a charging rate worth C/10 .

The calibration device ends each discharge phase 12.2 when a predetermined condition is met. For example, the calibration device ends the discharge phase 12.2 when the voltage across the battery cell 2 reaches a predetermined minimum voltage, strictly lower than the predetermined maximum voltage.

The predetermined minimum and maximum voltages depend in particular on the nature of the materials present within the battery element 2.

The predetermined minimum and maximum voltages are chosen so as to reach a limit lithiation state making it possible to preserve the insertion structure of the materials used in the battery element 2.

In addition, the electrolyte 8 is chosen to be stable at least between the predetermined minimum voltage and maximum voltage.

For example, the predetermined minimum and maximum voltages for a known Gr/NMC (i.e. graphite/LiNi _1/3 Mn _1/3 Co _1/3 O ₂ ) battery cell are respectively 2.7 V and 4.2 V. According to another example, the minimum and maximum predetermined voltages for a known Gr/LFP (i.e. graphite/LiFePO ₄ ) battery cell are respectively 2.0 V and 3.6 V.

Preferably, at least one charging phase 12.1 or at least one discharging phase 12.2 of the cyclic step 12 is followed by a stationary phase 12.4. During each stationary phase 12.4, no exchange of electrical energy occurs between the calibration device and the battery element 2. Each stationary phase has a predetermined duration, for example 30 minutes.

Advantageously, once the initialization step 10 and the plurality of cyclic steps 12 have been implemented for the battery element 2, the initialization step 10 and the plurality of cyclic steps 12 are implemented. implemented for another battery element 2 among the plurality of battery elements 2 of the same family which was initially supplied.

In this case, the calibration device uses the VAL _lim limit values obtained for various charging regimes, different initial surface capacities of battery elements 2 and different states of health to calculate, by regression, the parameters of a mathematical model linking the limit value VAL _lim to the charging regime, to the initial surface capacity Q ₀ of battery cell 2 and to the state of health of battery cell 2.

• où VAL_lim est la valeur limite de l'état de charge S, exprimée en pourcent ;
• Q₀ est la capacité surfacique initiale de l'élément de batterie 2, exprimée en milliampère heure par centimètre carré ;
• H est l'état de santé de l'élément de batterie 2, exprimé en pourcent ;
• C est le régime de charge ;
• log est l'opérateur logarithme décimal ; et
• a₀, a₁, a₂, a₃, a₁₂, a₂₃, a₁₁, a₂₂ et a₃₃ sont des coefficients réels obtenus par régression et dépendant de la famille de batteries étalonnée.

For example, the model is a quadratic model of the form: $${\mathrm{VAL}}_{\mathrm{lime}}={\mathrm{has}}_0+{\mathrm{has}}_1{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0+{\mathrm{has}}_2\log \left(\mathrm{VS}\right)+{\mathrm{has}}_3\mathrm{H}+{\mathrm{has}}_{12}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\log \left(\mathrm{VS}\right)+{\mathrm{has}}_{13}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\mathrm{H}+{\mathrm{has}}_{23}\log \left(\mathrm{VS}\right).\mathrm{H}+{\mathrm{has}}_{11}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0^2+{\mathrm{has}}_{22}\log {\left(\mathrm{VS}\right)}^2+{\mathrm{has}}_{33}{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">H</figure-callout>}}^2$$

• where VAL _lim is the limit value of the state of charge S, expressed in percent;
• Q ₀ is the initial surface capacity of battery cell 2, expressed in milliampere hours per square centimeter;
• H is the health status of battery cell 2, expressed in percent;
• This is the charging regime;
• log is the decimal logarithm operator; And
• a ₀ , a ₁ , a ₂ , a ₃ , a ₁₂ , a ₂₃ , a ₁₁ , a ₂₂ and a ₃₃ are real coefficients obtained by regression and depending on the calibrated battery family.

For example, from the data that made it possible to construct the curves of the Figure 4 (charging regime equal to C/10, initial surface capacity worth 7.61 mAh/cm ² ), the model obtained is written, for the limiting value of the state of charge S: $${\mathrm{VAL}}_{\mathrm{lime}}=-1.51-0.1970.{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0+21.2.\log \left(\mathrm{VS}\right)+0.3475.\mathrm{H}-0.866.\log \left(\mathrm{VS}\right).\mathrm{H}+20.51.\log {\left(\mathrm{<figure-callout\; id="2"\; label="VS"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">VS</figure-callout>}\right)}^2$$

Preferably, if the cyclic step 12 is implemented at different temperatures, so that the limit value VAL _lim takes values associated with different temperatures, the calibration device uses the limit values VAL _lim obtained for various temperatures, various charging regimes, different initial surface capacities of battery elements 2 and different states of health to calculate, by regression, the parameters of a mathematical model linking the limit value VAL _lim to the charging regime, to the temperature of the battery cell 2, to the initial surface capacity Q ₀ of battery cell 2 and to the health state of battery cell 2.

Alternatively, during the implementation of the calibration process, the predetermined characteristic quantity chosen is the voltage U between the cathode 4 and the anode 6 of the battery cell 2 chosen. In other words, the voltage U is taken as a quantity whose value is representative of the value of the electric potential of the anode 6.

In this case, the implementation of the calibration method differs from the implementation previously described in the following.

During each charging phase 12.1, the calibration device does not calculate the state of charge S of the battery cell 2, but measures the voltage U between the electrodes 4, 6 of the battery cell 2. The voltage U increases during charging phase 12.1.

In addition, during each charging phase 12.1, the calibration device determines a limit value VAL _lim of the voltage U for which the potential difference between the anode 6 and the reference electrode becomes less than or equal to one predetermined threshold.

In addition, during each charging phase 12.1, the calibration device records the limit value VAL _lim of the voltage U, the limit value VAL _lim being associated with the family of the battery element 2 chosen, with the capacity initial surface area of the chosen battery element 2, to the state of health of the chosen battery element and to the charging regime of the charging phase 12.1 considered.

Furthermore, the calibration device uses the VAL _lim limits of the voltage U obtained for various charging regimes and different initial surface capacities and different health states of battery cells to calculate, by regression, the parameters of a model mathematical linking the limit value VAL _lim of the voltage U to the charging regime and to the initial surface capacity Q ₀ and to the state of health H of the battery cell 2.

• où VAL_lim est la valeur limite de la tension, exprimée en volt ;
• Q₀ est la capacité surfacique initiale de l'élément de batterie 2, exprimée en milliampère heure par centimètre carré ;
• C est le régime de charge ;
• H est l'état de santé de l'élément de batterie 2, exprimé en pourcent ;
• log est l'opérateur logarithme décimal ; et
• b₀, b₁, b₂, b₃, b₁₃, b₂₃, b_11, b₂₂ et b₃₃ sont des coefficients réels obtenus par régression et dépendant de la famille de batteries étalonnée.

For example, the model is a quadratic model of the form: $${\mathrm{VAL}}_{\mathrm{lime}}={\mathrm{<figure-callout\; id="0"\; label="b"\; filenames="imgf0002.png"\; state="\{\{state\}\}">b</figure-callout>}}_0+{\mathrm{b}}_1{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0+{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">b</figure-callout>}}_2\log \left(\mathrm{VS}\right)+{\mathrm{b}}_3\mathrm{H}+{\mathrm{b}}_{12}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\log \left(\mathrm{VS}\right)+{\mathrm{b}}_{13}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0.\mathrm{H}+{\mathrm{b}}_{23}\log \left(\mathrm{VS}\right).\mathrm{H}+{\mathrm{b}}_{11}{\mathrm{<figure-callout\; id="0"\; label="Q"\; filenames="imgf0002.png"\; state="\{\{state\}\}">Q</figure-callout>}}_0^2+{\mathrm{<figure-callout\; id="22"\; label="b"\; filenames="imgf0003.png"\; state="\{\{state\}\}">b</figure-callout>}}_{22}\log {\left(\mathrm{VS}\right)}^2+{\mathrm{b}}_{33}{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="imgf0001.png,imgf0002.png"\; state="\{\{state\}\}">H</figure-callout>}}^2$$

• where VAL _lim is the voltage limit value, expressed in volts;
• Q ₀ is the initial surface capacity of battery cell 2, expressed in milliampere hours per square centimeter;
• This is the charging regime;
• H is the health status of battery cell 2, expressed in percent;
• log is the decimal logarithm operator; And
• b ₀ , b ₁ , b ₂ , b ₃ , b ₁₃ , b ₂₃ , b _11, b ₂₂ and b ₃₃ are real coefficients obtained by regression and depending on the calibrated battery family.

Preferably, if the cyclic step 12 is implemented at different temperatures, so that the limit value VAL _lim of the voltage U takes values associated with different temperatures, the calibration device uses the limit values VAL _lim obtained for various temperatures, various charging regimes, different initial surface capacities of battery elements 2 and different states of health to calculate, by regression, the parameters of a mathematical model linking the limit value VAL _lim of the voltage to the charging regime , at the temperature of battery cell 2, at the initial surface capacity Q ₀ of battery cell 2 and at the state of health H of battery cell 2.

A charging device 20 according to the invention, connected to a battery 200 to be charged, is shown on the Figure 5 .

The battery 200 is obtained by placing identical battery elements 2 in series and/or in parallel.

The charging device 20 comprises a power supply circuit 22 and a computer 26.

Preferably, the charging device 20 also includes at least one temperature sensor 24, each temperature sensor 24 being configured to measure the temperature of the battery 200.

The power supply circuit 22 is configured to transmit electrical energy to the battery 200, from an electrical energy source 28. More precisely, the power supply circuit 22 is configured to transmit electrical energy to the battery 200 at a predetermined charging regime.

The calculator 26 is configured to calculate the value of the predetermined characteristic quantity relating to each battery element 2 of the battery 200 over time. For example, the calculator 26 is configured to calculate the state of charge of each battery element 2 of the battery 200 over time. Alternatively, the calculator 26 is configured to calculate the voltage between the electrodes 4, 6 of each battery element 2 of the battery 200 over time.

The calculator 26 is also configured to store, for at least one family of battery elements 2, the model of the limit value VAL _lim , for the predetermined characteristic quantity, which was obtained according to the calibration method as described previously .

The calculator 26 is also configured to evaluate the state of health H of each battery element 2 of the battery 200.

In addition, the calculator 26 is configured to calculate the limit value VAL _lim of the predetermined characteristic quantity chosen for each battery element 2 of the battery 200 to be charged, as a function of the family of battery elements 2, of the initial surface capacity of the battery elements 2, the state of health H evaluated for each battery element 2 and the charging regime according to which the power supply circuit 22 supplies electrical energy to the battery 200.

Furthermore, the computer 26 is configured to detect a situation, called a “limit state”, in which the value of the predetermined characteristic quantity relating to at least one battery element 2 reaches a predetermined fraction of the corresponding limit value VAL _lim .

The computer 26 is also configured to control the charging regime according to which the power supply circuit 22 supplies electrical energy to the battery 200.

In particular, the computer 26 is configured to control, in the event of detection of the limit state, the reduction of the charging regime according to which the power supply circuit 22 supplies electrical energy to the battery 200.

The charging of the battery 200 by means of the charging device 20 is described with reference to the Figure 6 .

During a configuration step 30, a battery 200 comprising identical battery elements 2 of a given family of Li-ion batteries, and having a given initial surface capacity Q ₀ and a given state of health H, is provided.

An operator provides information on the architecture of the battery 200 in the charging device 20. In particular, the operator provides information on how the battery elements 2 of the battery 200 are connected to each other.

The operator also informs, in the charging device 20, the family and the initial surface capacity of the battery elements 2 of the battery 200.

Then, the charging device 20 evaluates the state of health of each battery element 2 of the battery 200.

Then, the charging device 20 loads the mathematical model of the limit charge VAL _lim for the family corresponding to the battery elements 2 of the battery 200.

Alternatively, the charging device 20 is intended to charge predetermined batteries 200, comprising identical battery elements 2. In this case, operator intervention during the configuration step 30 is not required, the information relating to the family and the initial surface capacity of the battery elements 2 of the battery 200, as well as to the The architecture of the battery 200, being pre-recorded in the charging device 20.

According to another variant, the charging device 20 comprises a detector (not shown) configured to identify the battery 200, that is to say to determine the architecture of the battery elements 2, as well as the family and the surface capacity initial of the battery elements 2 of the battery 200. In this case, the intervention of the operator during the configuration step 30 is not required, the information relating to the family and the initial surface capacity of the battery elements 2 of the battery 200, as well as the architecture of the battery 200, being pre-recorded in the charging device 20.

Then, during a power supply step 32, the power supply circuit 22 supplies electrical energy to the battery 200, at a given charging rate.

Furthermore, during the power supply step 32, each temperature sensor 24 preferably measures the temperature of the battery 200.

During the power supply step 32, the calculator 26 calculates the limit state of charge VAL _lim for the family and the initial surface capacity of the battery elements 2, the state of health of each battery element 2 and the speed charging according to which the power supply circuit 22 supplies electrical energy to the battery 200. Advantageously, the calculator 26 also takes into account the temperature measured by each temperature sensor 24, if present, for the calculation of the state of charge limit VAL _lim .

In addition, during the power supply step 32, the calculator 26 calculates the state of charge of each battery element 2.

If the state of charge of a battery element 2 of the battery 200 reaches the predetermined fraction of the limit value VAL _lim for the state of charge S, then the calculator 26 detects a limit state.

Alternatively, the calculator 26 calculates the voltage between the electrodes 4, 6 of each battery element 2. In this case, if the voltage between the electrodes of a battery element 2 of the battery 200 reaches the predetermined fraction of the value limit VAL _lim for the voltage U, then the computer 26 detects a limit state.

Alternatively, the calculator 26 evaluates the state of health of each battery element 2 throughout the power supply step 32, the value of the limit state of charge (or the limit voltage) being updated, over time, with the evolution of the state of health.

Preferably, in the event of detection of a limit state, during a reconfiguration step 34 successive to the power supply step, the computer 26 commands the power supply circuit 22 to select a new charging regime distinct from the charging speed of the previous feeding step 32, preferably strictly lower than the charging speed of the previous feeding step 32.

For example, in the case where the predetermined characteristic quantity is the state of charge S, during the reconfiguration step 34, the computer 26 commands the power supply circuit 22 to select a new charging speed strictly lower than the charging speed of the previous power supply step 32.

The step successive to the reconfiguration step 34 is the power supply step 32, the power supply circuit 22 supplying electrical energy to the battery 200 according to the new charging regime.

Claims (14)

1. Method for calibrating a family of Li-ion battery elements (2), characterised in that it includes the steps of:

   - providing a battery element (2) of the family, the battery element (2) having a corresponding given initial surface area capacity;

   - at least one evaluation of a current state of health of the battery element (2) supplied;

   - for the battery element (2) supplied, implementing a plurality of successive cycles, each comprising a charging phase (12.1) and a discharging phase (12.2), at least two charging phases (12.1) being carried out at different charging rates;

   - for each charging phase:

   • measuring an electrical potential of an anode (6) of the battery element (2) supplied;

   • calculating the value of a predetermined characteristic variable of the battery element (2);

   • determining a limit value of the predetermined characteristic variable for which the potential of the anode (6) becomes less than or equal to a predetermined threshold;

   • recording, in a memory, the limit value for said family the said initial surface area capacity, the charging rate of said charging phase (12.1) and the evaluated state of health.
2. Calibration method according to claim 1, wherein each evaluation of the current state of health of the battery element (2) supplied is carried out during a corresponding cycle.
3. Calibration method according to claim 1 or 2, including calculating, by regression, a mathematical model relating the limit value of the predetermined characteristic variable to the charging rate.
4. Calibration method according to claim 3, wherein the mathematical model relates the limit value of the predetermined characteristic variable to the charging rate, to the initial surface area capacity of the battery element (2) supplied and to the state of health of the battery element (2) supplied.
5. Calibration method according to claim 4, wherein the mathematical model is a quadratic model of the form: $$\begin{array}{l}{\mathrm{VAL}}_{\lim }={\mathrm{a}}_0+{\mathrm{a}}_1{\mathrm{Q}}_0+{\mathrm{a}}_2\log \left(\mathrm{C}\right)+{\mathrm{a}}_3\mathrm{H}+{\mathrm{a}}_{12}{\mathrm{Q}}_0.\log \left(\mathrm{C}\right)+{\mathrm{a}}_{13}{\mathrm{Q}}_0.\mathrm{H}+{\mathrm{a}}_{23}\log \left(\mathrm{C}\right).\mathrm{H}+{\mathrm{a}}_{11}{\mathrm{Q}}_0^2+\\ {\mathrm{a}}_{22}\log {\left(\mathrm{C}\right)}^2+{\mathrm{a}}_{33}{\mathrm{H}}^2\end{array}$$

   

   where VAL_lim is the limit value of the predetermined characteristic variable;

   Q₀ is the initial surface area capacity of the battery element (2);

   H is the state of health of the battery element (2);

   C is the charging rate;

   log is the decimal logarithm operator; and

   a₀, a₁, a₂, a₃, a₁₃, a₂₃, a₁₁, a₂₂ and a₃₃ are real coefficients.
6. Calibration method according to claim 4, wherein, for each cycle, the battery element (2) is kept at a constant temperature, at least two charging phases (12.1) being carried out at different temperatures,

   the limit value recorded for each charging phase being associated with the corresponding temperature,

   the calculated mathematical model relating the limit value of the predetermined characteristic variable to the charging rate, to the initial surface area capacity of the battery element (2) supplied, to the state of health of the battery element (2) supplied as well as to the temperature of the battery element (2) supplied.
7. Calibration method according to any one of claims 1 to 6, wherein the potential of the anode (6) is calculated from the potential difference between the anode (6) and a reference electrode.
8. Calibration method according to claim 7, wherein the reference electrode is an Li + /Li electrode, the limit value of the predetermined characteristic variable being reached when the potential difference between the anode (6) and the reference electrode is zero.
9. Calibration method according to any one of claims 1 to 8, wherein the predetermined characteristic variable is a state of charge of the battery element (2), or a voltage between a cathode (4) and the anode (6) of the battery element (2).
10. Charging method including the steps of:

    - providing a battery (200) comprising at least one Li-ion battery element (2) belonging to a family of Li-ion battery cells and having a given initial surface area capacity;

    - evaluating a state of health of each battery element (2) of the battery (200);

    - supplying the battery (200) supplied with electrical energy at a given charging rate;

    - for each battery element (2), calculating a limit value of a corresponding predetermined characteristic variable, for said family, said initial surface area capacity, said state of health and said charging rate;

    - for each battery element (2), calculating the value of the predetermined characteristic variable; and

    - detecting a limit state when, for at least one battery element (2), the value of the predetermined characteristic variable reaches a predetermined fraction of the corresponding limit value,

    wherein the limit value is obtained by applying the calibration method according to any one of claims 1 to 9.
11. Method according to claim 10, further including reducing the charging rate on detection of the limit state.
12. Computer program product comprising program code instructions which, when executed on a computer, implement the steps of evaluating, calculating the limit value, calculating the value of the predetermined characteristic variable and detecting the method according to claim 10.
13. Charging device including:

    - a supply circuit configured to transmit electrical energy to an Li-ion battery (200), at a given charging rate, from a source (28) of electrical energy;

    - a computer (26) configured to:

    • evaluate a state of health of each battery element (2) of the battery (200);

    • calculate the value of a predetermined characteristic variable relating to at least one battery element (2) of the battery (200) over time;

    • detect a limit state when the value of the predetermined characteristic variable relating to at least one battery element (2) reaches a predetermined fraction of a limit value, the limit value being a function of a family and an initial surface area capacity of the battery element (2), the evaluated state of health and the charging rate.
14. Device according to claim 13, wherein the computer (26) is further configured to control the reduction of the charging rate in the event of detection of the limit state.

Images